You are on page 1of 13

CHAPTER 3

FORCED OSCILLATIONS
Reference: George C. Kings, Vibrations and waves, A John Wiley and Sons, Ltd., Publication, 2009.
In this lecture we will concentrate our attention to forced oscillations where we apply a periodic
driving force to the system. We are surrounded by examples of such forced oscillations.
Before discussing forced harmonic oscillator let us remember the properties and applications
of damped harmonic oscillator described in the previous chapter.
Review of Damped Harmonic Oscillator
The mechanical example of a mass moving through a viscous
fluid is shown in figure and we saw that the fluid offered a
resistance that damped the motion.
The equation of motion for damped motion is in the form:

𝑑2 𝑥 𝑑𝑥
+ 𝛾 + 𝜔02 𝑥 = 0.
𝑑𝑡 2 𝑑𝑡
𝑘 𝑏
𝑤𝑕𝑒𝑟𝑒 𝜔02 = ; 𝛾 =
𝑚 𝑚
The general solution of the equation are given by
𝛾
𝑥 = 𝑥0 𝑒 −2 𝑡 cos 𝜔𝑡 + 𝜙
where

𝛾2
𝜔= 𝜔02 −
4
Depending on the relation between
damping constant and natural frequency
we have three cases:

(i) light damping; for the condition
𝛾
2
< 𝜔0 .
(ii) heavy or over damping; for the
𝛾
condition, 2 > 𝜔0 .
(iii) critical damping; for the condition
𝛾
2
= 𝜔0 .

The motion of a damped oscillator for the cases of light damping, heavy damping and critical
damping.is illustrated in figure.
The mechanical energy of a damped harmonic oscillator is eventually dissipated as heat and the energy
of the system is found in the form
1
𝐸 = 𝑚𝜔02 𝑥0 𝑒 −𝛾𝑡 = 𝐸0 𝑒 −𝛾𝑡
2

Defining = 1/𝛾 . The reciprocal of 𝛾 is the time taken for the energy of the oscillator to reduce by a factor of 𝑒. İt seen that the energy of the oscillator decays exponentially with time as shown in Figure.5 Transient Phenomena 3. although still with the same frequency. . Power dissipated by resistive or damped or drag force is 𝑑𝐸 𝑃= = −𝑏𝑣 2 𝑑𝑡 Since 𝑃 = 𝐹𝑣 The quality factor Q of a damped harmonic oscillator is defined as 𝜔0 𝑄= .4 Resonance in Electrical Circuits 3. We get the largest amplitude at resonance because this is the frequency at which the pendulum‘wants’ to oscillate. The energy of an oscillator is dissipated because it does work against the damping force at the rate (𝐹𝑑 = −𝑏𝑣). along a horizontal direction. we obtain 𝑡 𝐸 = 𝐸0 𝑒 −𝜏 where 𝜏 has the dimensions of time and is called the decay time or time constant of the system. As the driving frequency is increased further the amplitude of oscillation decreases but perhaps more surprisingly the mass now moves in the opposite direction to the point of suspension.6. but all forced oscillators behave in this manner. At very low driving frequencies the pendulum mass closely follows the movement of the point of suspension with them both moving in the same direction as each other. As the driving frequency is increased the amplitude of oscillation increases dramatically and becomes much larger than the movement of the point of suspension. We drive the pendulum by moving its point of suspension backwards and forwards harmonically. The system is then said to be in resonance.1 PHYSICAL CHARACTERISTICS OF FORCED HARMONIC MOTION We can observe the main physical characteristics of forced harmonic motion using a simple pendulum.2 Forced oscillations with damping 3. We might rightly suspect that the maximum amplitude occurs when the pendulum is driven close to its natural frequency of oscillation.2 Physical Characteristics of Forced Harmonic Motion 3.e. The simple pendulum serves as a useful example.3 Power Absorbed During Forced Oscillations 3.1 Complex numbers 3.where 𝐸0 is the total energy of the oscillator at t = 0.3 Use of the complex representation for forced oscillations with damping 3.they have the same amplitude and move in phase.2 The use of complex numbers to represent physical quantities 3.1 The Equation of Motion of a Forced Harmonic Oscillator 3. Remember physical meaning of the 𝜔0 𝑎𝑛𝑑 𝛾 then we define 𝑄-value 𝐸𝑛𝑒𝑟𝑔𝑦 𝑠𝑡𝑜𝑟𝑒𝑑 𝑖𝑛 𝑡𝑕𝑒 𝑜𝑠𝑐𝑖𝑙𝑙𝑎𝑡𝑜𝑟 𝑄= 𝐸𝑛𝑒𝑟𝑔𝑦 𝑑𝑖𝑠𝑠𝑖𝑝𝑎𝑡𝑒𝑑 𝑝𝑒𝑟 𝑟𝑎𝑑𝑖𝑎𝑛 Outline of this chapter 3. At even higher frequencies we reach the situation where the pendulum mass hardly moves at all.6 The Complex Representation of Oscillatory Motion 3.2. 𝛾 Note that 𝜔0 𝑎𝑛𝑑 𝛾 have the same dimensions therefore 𝑄 is a pure number.6. This is because it has inertia.1 Undamped forced oscillations 3.6. i.2.

The spring constant is k and the mass moves without friction on a horizontal surface.1 Undamped forced oscillations We begin with a mass m on a horizontal spring as shown in Figure. i. The displacement x is then equal to F0/k. namely ω = 0. Here we move the upper end s of the spring up and down harmonically in the vertical direction according to ξ = a cos ωt where a is the amplitude and ω is the applied frequency.2 THE EQUATION OF MOTION OF A FORCED HARMONIC OSCILLATOR 3.e. If the oscillator is driven from its equilibrium position and then left to itself. The displacement x is measured from the equilibrium position of the mass. So. A periodic driving force will. We shall deduce a solution for above equation and see how the oscillator behaves as we change the angular frequency ω of the driving force. at very low driving frequencies when ω tends to zero. We deduce a solution for forced oscillations using the arrangement of a mass m on a vertical spring. the acceleration term is zero. Before we discuss this differential equation of motion in detail. it will oscillate with its natural frequency 0.2. ω = 0. (This simple but very informative experiment can be performed using a few elastic bands strung together with a small mass attached to the lower end. If we have a force F0 that does not change with time. From Newton’s second law we obtain 𝑑2 𝑥 𝑚 2 + 𝑘𝑥 = 𝐹0 𝑐𝑜𝑠𝜔𝑡 𝑑𝑡 This is the equation of motion for forced oscillations of a harmonic oscillator when there is no damping. therefore. The mass is acted upon by the combination of the restoring force from the spring and the applied driving force. the amplitude of oscillation will tend to the value F0/k. The mathematically complete solution of above equation is indeed a simple sum of these two motions. This system is similar to the one described in chapter I but now we imagine that a periodic driving force F = F0cos ωt is applied to it. however. We must expect. of spring constant k. let us consider the situation qualitatively. as shown below figure. that the actual motion in this case is some kind of a superposition of oscillations at the two frequencies  and 0. First we note one limit of ω. try to impose its own frequency  on the oscillator.) .3.

𝑑2 𝑥 𝑚 2 + 𝑘𝑥 = 𝑘𝑎𝑐𝑜𝑠𝑡 𝑑𝑡 where 𝑘𝑎 = 𝐹0 At very low frequencies the amplitude of oscillations tends to the value of the amplitude a of the point of suspension. 𝜔2 1− 2 𝜔 0 When δ = π in the first equation we obtain . Substituting x and its second derivative indifferential equation and 𝑘 using 𝜔2 = 𝑚 gives −𝜔2 𝐴 𝜔 cos 𝜔𝑡 − 𝛿 + 𝜔02 𝐴 𝜔 cos 𝜔𝑡 − 𝛿 = 𝜔02 𝑎 cos 𝜔𝑡. however. The minus sign of δ in the equation implies that the displacement lags behind the driving force and this is indeed the case in forced oscillations. we get tan δ = 0 and so δ = 0 or π. When δ = 0 in the first equation we obtain 𝑎 𝐴 𝜔 = 𝑓𝑜𝑟 𝜔 > 𝜔0 to get positive 𝐴 𝜔 . As ω is increased above the resonance frequency the amplitude decreases and the mass moves in the opposite direction to the driving force.We measure the displacement x from the equilibrium position of the mass and take displacements in the downward direction as positive. the mass moves up and down periodically at the same frequency ω as the driving force. From our previous considerations. Expanding terms in cos(ωt − δ) leads to −𝜔2 𝐴 𝜔 (cos 𝜔𝑡 𝑐𝑜𝑠 𝛿 + 𝑠𝑖𝑛𝜔𝑡𝑠𝑖𝑛𝛿) + 𝜔02 𝐴 𝜔 (cos 𝜔𝑡 𝑐𝑜𝑠 𝛿 + 𝑠𝑖𝑛𝜔𝑡𝑠𝑖𝑛𝛿) = 𝜔02 𝑎 cos 𝜔𝑡 Then equating coefficients of cos ωt and sin ωt we obtain 𝜔2 𝜔2 𝐴 𝜔 = 1− 𝑐𝑜𝑠 𝛿 = 𝑎 𝑎𝑛𝑑 𝐴 𝜔 = 1 − 𝑠𝑖𝑛 𝛿 = 0 𝜔02 𝜔02 Dividing second equation to first. At all frequencies. The resultant equation of motion is 𝑑2 𝑥 𝑚 2 = −𝑘(𝑥 − 𝜉) 𝑑𝑡 substituting for 𝜉. This behaviour suggests that the displacement x of the mass can be written as 𝑥 = 𝐴 𝜔 cos⁡ (𝜔𝑡 − 𝛿) In this equation. The phase angle  relates to the initial conditions and along with A completely defines the free oscillations and δ is the phase angle between the driving force and the resultant displacement. A(ω) is the physical amplitude that we observe and which we naturally define as a positive quantity. δ is a phase angle but has a different meaning to the phase angle  in the expression given in Chapter 1. Under these conditions the motion is governed by the spring constant or stiffness of the spring. the amplitude tends to zero when the motion is governed by the inertia of the mass. As ω is increased the amplitude of oscillation increases dramatically as the resonance frequency is approached. we expect δ to be zero at very low frequencies and equal to π at very high frequencies. At higher frequencies still.

as well as the amplitude A(ω). The behaviour of A(ω) as ω approaches ωo is clearly unphysical and arises because damping forces have been neglected. is shown in Figure (a) and the behaviour of the phase angle δ with respect to ω is shown in Figure (b). and δ → π. tan δ →0. to obtain 𝑑2 𝑥 𝑑𝑥 𝐹0 2 +𝛾 + 𝜔02 𝑥 = 𝑐𝑜𝑠𝑡 𝑑𝑡 𝑑𝑡 𝑚 This is the equation for forced oscillations with damping. Again we try a solution of the form x = A(ω) cos(ωt − δ) and substitute for x and its derivatives in the equation. tan δ =∞. and δ → 0. This is again because the effects of damping have not been included. The change of the phase angle δ from zero to π is consistent with the behaviour of the forced oscillators we have considered but its sharp and abrupt change at ω = ωo is unphysical. remembering that F0 = ka. as ω→∞. 3. Using the mathematical nomenclature → meaning ‘tends to’. A plot of A(ω) against ω.2.2 Forced oscillations with damping We will again assume that the damping force is directly proportional to the velocity of the mass as we did in chapter 2. according to above equations. 𝜔2 1− 2 𝜔 0 It is seen that A(ω) tends to a(= F0/k) as ω tends to zero. depends on the driving frequency ω. Adding this term to the differential equation given in previos section we get 𝑑2 𝑥 𝑑𝑥 𝑚 2 +𝑏 + 𝑘𝑥 = 𝐹0 𝑐𝑜𝑠𝑡 𝑑𝑡 𝑑𝑡 We make the substitutions b/m = γ and k/m = 𝜔02 . Then equating the coefficients of cos ωt and sin ωt we obtain 𝐴 𝜔 𝜔02 − 𝜔2 𝑐𝑜𝑠 𝛿 + 𝛾𝑠𝑖𝑛𝛿 = 𝜔02 𝑎 and 𝜔02 − 𝜔2 𝑠𝑖𝑛𝛿=𝛾𝑐𝑜𝑠𝛿 giving 𝛾 𝑡𝑎𝑛𝛿 = 2 𝜔0 − 𝜔 2 We see that the phase angle δ. tan δ →1/(−∞). and when ω = ωo. and δ = π/2 From last equation we can construct the right-angled triangle shown in Figure to obtain 𝛾 𝑠𝑖𝑛𝛿 = 1 [ 𝜔02 − 𝜔 2 2 + 𝜔 2 𝛾 2 ]2 . These are present in real systems and limit the maximum value of A(ω). inspection of Equation shows that as ω → 0. −𝑎 𝐴 𝜔 = 𝑓𝑜𝑟 𝜔 < 𝜔0 to get positive 𝐴 𝜔 .

𝜔02 − 𝜔2 𝑐𝑜𝑠𝛿 = 1 [ 𝜔02 − 𝜔 2 2 + 𝜔 2 𝛾 2 ]2 Substituting for sin δ and cos δ in related above equation we finally obtain 𝑎𝜔02 𝐴 𝜔 = 1 [ 𝜔02 − 𝜔 2 2 + 𝜔 2 𝛾 2 ]2 which describes the amplitude dependence on driving frequency ω for forced oscillations with damping. . A(ω)→a(=F0/k). when γ is zero. We note that this equation reduces to the result for the undamped case. Finally. This occurs when 𝑑 1 [ 𝜔02 − 𝜔2 2 + 𝜔2 𝛾 2 ]2 = 0 𝑑𝜔 from which 1 𝛾2 2 𝜔 = 𝜔0 1 − = 𝜔𝑚𝑎𝑥 2𝜔02 We can find the maximum value of the amplitude Amax by substituting ωmax 𝑎𝜔0 𝛾 𝐴𝑚𝑎𝑥 = 𝛾 2 12 (1 − ) 4𝜔02 The dependences of the amplitude A(ω) and the phase angle δ on the driving frequency ω are shown in Figure . the phase angle varies continuously. A(ω)= aωo/γ. the maximum amplitude remains finite although large and occurs at a lower frequency than ωo. as ω→∞. using the substitution F0 = ka. Furthermore. (We recall that δ is the phase angle by which the displacement lags behind the driving force. the denominator in equation must be a minimum. These results are similar to the undamped case except that the amplitude does not go to infinity at ω = ωo. A(ω)→0. we can write amplitude. Inspection of this equation shows that as ω → 0. however. the maximum amplitude of oscillation no longer occurs at ωo.) These curves are similar to those for the undamped case With damping. For A(ω) to be a maximum. and when ω = ωo.

𝐹0 𝑚 𝐴 𝜔 = 1 [ 𝜔02 − 𝜔 2 2 + 𝜔 2 𝛾 2 ]2 3. but if there is no damping there is no dissipation of energy. The power absorbed by the oscillator to sustain its motion is exactly equal to the rate at which the energy is dissipated . . as 𝜔0 𝜔 𝜔02 − 𝜔2 = − (𝜔0 𝜔) 𝜔 𝜔0 We obtain 𝑎𝜔02 𝑣0 𝜔 = 1 𝜔 𝜔 2 2 [ 𝜔0 − 𝜔 𝜔0 + 𝛾 2 ]2 0 It is seen that as ω → 0. During steady state oscillations. as ω→∞. 𝑣0 𝜔 → 0.3 POWER ABSORBED DURING FORCED OSCILLATIONS In Section 3. However. As usual we will assume that the damping force is proportional to the velocity of the mass and so we begin by considering how the velocity varies during forced oscillations.2. and the value of 𝑣0(ω) passes through a maximum at exactly ω = ωo. energy must be provided to stretch or compress the spring but this energy is recovered as the spring returns to its equilibrium length. The displacement x of the mass is given by 𝑥 = 𝐴 𝜔 cos⁡ (𝜔𝑡 − 𝛿) where 𝑎𝜔02 𝐴 𝜔 = 1 [ 𝜔02 − 𝜔 2 2 + 𝜔 2 𝛾 2 ]2 and so velocity is given by 𝑑𝑥 𝑣= = −𝑣0 𝜔 sin 𝜔𝑡 − 𝛿 𝑑𝑡 where 𝑣0 𝜔 = 𝜔𝐴 𝜔 Substituting for A(ω) in last equation gives 𝑎𝜔02 𝜔 𝑣0 𝜔 = 1 [ 𝜔02 − 𝜔 2 2 + 𝜔 2 𝛾 2 ]2 Rewriting (𝜔02 − 𝜔2 ). 𝑣0 𝜔 → 0. Consequently. when it is equal to aω2/γ. a real oscillator loses energy because of the frictional damping forces that are invariably present. for the ideal situation where there is no damping. The applied force drives the mass back and forth.1 we described the application of a periodic driving force to a mass on the end of a spring. the total power delivered to the oscillator over each complete cycle is zero. The driving force has to restore this lost energy.

Since the damping force and the velocity are time-dependent. we can replace ω by ωo everywhere in last equation except in the term (𝜔02 − 𝜔2 ) which is replaced by 𝜔02 − 𝜔2 = 𝜔0 + 𝜔 𝜔0 − 𝜔 ≈ 2𝜔0 −∆𝜔 . i. where _ω ≡ ω − ωo. An example of such a power resonance curve is shown in Figure.e. It is seen from the figure. 𝜔02 = k/m and a =F0/k. 2 Substituting equation for 𝑣0 𝜔 and using b = mγ. we get our final expression for the power resonance curve: 𝐹02 𝑃 𝜔 = . as ω →0. 2𝑚[ 𝜔02 − 𝜔 2 2 + 𝜔 2 𝛾 2 ] A plot of P(ω) against ω gives the power resonance curve of the oscillator. When the driving frequency is close to the resonance frequency ωo. as ω→∞. 4∆𝜔 2 2𝑚𝛾 +1 𝛾2 The maximum value of P(ω) is given by . ω ≈ ωo. P(ω) →0 and the maximum value of P(ω) occurs exactly when ω = ωo.The rate of energy loss due to damping is equal to the damping force times the velocity of the mass. P(ω) is given by 𝑡 0+𝑇 1 𝑃 𝜔 = 𝑃 𝑡 𝑑𝑡. where T is the period. This width characterises the sharpness of the response of the oscillator to an applied force. 𝑇 𝑡0 Thus 𝑡 0+𝑇 2 𝑏 𝑣0 𝜔 𝑃 𝜔 = 𝑠𝑖𝑛2 𝜔𝑡 − 𝛿 𝑑𝑡 𝑇 𝑡0 𝑏 𝑣0 𝜔 2 𝑃 𝜔 = . An important parameter of a power resonance curve is its full width at half height ωfwhh (see Figure ). Furthermore. since the instantaneous power varies it is more appropriate to talk in terms of the average power P(ω) absorbed over a complete cycle of oscillation between times to and to + T. P(ω) →0. we obtain 𝜔2 𝐹02 𝑃 𝜔 = . which shows how the power absorbed by the oscillator varies with the driving frequency. we must define the instantaneous power absorbed at time t by 𝑃 𝑡 = 𝑏𝑣 𝑡 𝑋𝑣 𝑡 Substituting 𝑣 𝑡 we get 𝑃 𝑡 = 𝑏 𝑣0 𝜔 2 𝑠𝑖𝑛2 𝜔𝑡 − 𝛿 . With these approximations.

When an atom is bathed in radiation it may under certain circumstances absorb this radiation. the atom will only absorb energy over a narrow range of frequencies close to the resonance frequency. plays the role of the driving force F0cos ωt. Applying Kirchoff’s law to the circuit gives the equation 𝑑2𝑞 𝑑𝑞 𝑞 𝐿 𝑑𝑡 2 + 𝑅 𝑑𝑡 + 𝑐 = 𝑉0 𝑐𝑜𝑠𝜔𝑡. 𝛾 𝜔𝑓𝑤 𝑕𝑕 𝑓𝑢𝑙𝑙 𝑤𝑖𝑑𝑡𝑕 𝑎𝑡 𝑕𝑎𝑙𝑓 𝑕𝑒𝑖𝑔𝑕𝑡 𝑜𝑓 𝑝𝑜𝑤𝑒𝑟 𝑐𝑢𝑟𝑣𝑒 This relationship offers a convenient way to measure the quality factor of an oscillator. An example of a resonance circuit is shown in Figure. where Q is quality factor defined in chapter 2. the oscillating electric field of the radiationinteracts with the atom which behaves like a forced oscillator. occur when 2∆𝜔 /γ = 1.e. in atomic and nuclear physics.4 RESONANCE IN ELECTRICAL CIRCUITS The phenomenon of resonance is also of great importance in electrical circuits. we can rewrite 𝐹02 𝑃 𝜔 = . they show up. It consists of an inductor L. ∆𝜔 1 2𝑚𝜔0 𝑄[4( 𝜔 )2 + 2 ] 0 𝑄 Power resonance curves for various values of the quality factor Q are presented in Figure . The half heights of the curve. Thus the full width at half height ωfwhh of the resonance curve is given by ωfwhh = ∆𝜔 = γ = ωo/Q. i. Since there is resistance in the circuit we are dealing with forced oscillations with damping. As for any oscillator of high Q. Using the relationship γ = ωo/Q. i. V0cos ωt. a capacitor C and a resistor R connected in series.e. when 2∆𝜔 = γ . 𝐹02 𝑃𝑚𝑎𝑥 = 2𝑚𝛾 and occurs when ∆𝜔 = 0. the curves are symmetric about their maxima except for low Q values. which are driven by an alternative (AC) voltage. Moreover. for example. equal to 𝑃𝑚𝑎𝑥 /2. Using the definition it can be written 𝜔0 𝜔0 𝑟𝑒𝑠𝑜𝑛𝑎𝑛𝑐𝑒 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑄= = = . Comparing this with 𝑑2 𝑥 𝑑𝑥 𝑚 2 +𝑏 + 𝑘𝑥 = 𝐹0 𝑐𝑜𝑠𝑡 𝑑𝑡 𝑑𝑡 we see that the alternating voltage. Power resonance curves are common in physical situations. and that m. exactly at ω = ωo.We see that the higher the value of Q the narrower is the power resonance curve. 3. Apart from mechanical and electrical systems. b and k for . V (t) = V0cos ωt. In a classical picture.

this is essentially the natural frequency ωo.the mechanical system are replaced by L. R and 1/C for the electrical system.𝑄 = = .𝛾 = . i. before it settles down to the steady state. at the resonance frequency and has the value V0/R. As indicated at the beginning of this chapter.e. this is not the whole story. 𝐿𝐶 𝐿 𝛾 𝑅 𝐶 Again using analogy the general solution is 𝑞 = 𝑞0 (𝜔)cos⁡𝜔𝑡 − 𝛿 where 𝑉0 𝑞0 (𝜔) = 𝐿 𝑅𝜔 2 1 [ 𝜔02 − 𝜔 2 2 + 𝐿 ]2 𝑉0 𝑞0 (𝜔) = 2 1 . Corresponding replacements give us 1 𝑅 𝜔0 1 𝐿 𝜔02 = . respectively. We can see this mathematically as follows. 1 2 𝜔[ 𝜔𝐶 − 𝜔𝐿 + 𝑅 ]2 where we have used 𝜔02 = 1/LC. 1 2 𝜔[ 𝜔𝐶 − 𝜔𝐿 + 𝑅 ]2 The maximum current amplitude in the circuit will occur when ω2 =𝜔02 . 3. However. During this initial period we thus have the sum of two oscillations of frequencies ω and ωo. as in the case of damped free oscillations . The initial behaviour of the oscillator. The rate at which they do this depends on the degree of damping. For the case of light damping. When the driving force is first applied and the system is disturbed from its equilibrium position the system will be inclined to oscillate at the frequency of its free oscillations. is referred to as its transient response. The current I flowing in the circuit is given by 𝑑𝑞 𝐼= = −𝑞0 (𝜔)𝜔sin 𝜔𝑡 − 𝛿 𝑑𝑡 −𝑉0 sin 𝜔𝑡 − 𝛿 = 2 1 . The equation for damped forced oscillations is 𝑑2 𝑥 𝑑𝑥 𝐹0 2 +𝛾 + 𝜔02 𝑥 = 𝑐𝑜𝑠𝑡 𝑑𝑡 𝑑𝑡 𝑚 If x1 is a solution of this equation then 𝑑 2 𝑥1 𝑑𝑥1 𝐹0 2 +𝛾 + 𝜔02 𝑥1 = 𝑐𝑜𝑠𝑡 𝑑𝑡 𝑑𝑡 𝑚 The equation for damped free oscillations is 𝑑2 𝑥 𝑑𝑥 2 +𝛾 + 𝜔02 𝑥 = 0 𝑑𝑡 𝑑𝑡 If x2 is a solution of this equation then . The system is then left oscillating at the frequency of the applied force and this is the steady state condition.5 TRANSIENT PHENOMENA Our discussion so far has emphasised that the oscillation frequency of a forced oscillator is the same as the frequency ω of the applied driving force. the oscillations of frequency ωo die away.

6 THE COMPLEX REPRESENTATION OF OSCILLATORY MOTION Oscillatory motion can also be described using complex numbers. We see that a complex number z has two components. Thus the complex conjugate of z = x + iy is z∗ = x − iy . An example of forced oscillations that start at time t = 0 is shown in Figure. a real part x and an imaginary part y. Analogous effects occur in AC circuits. as we shall see. This provides an elegant and concise representation and has important advantages. It is obtained by changing i to −i throughout. When the AC voltage is first applied to the circuit there will be a transient response. often denoted. A frequently useful quantity of a complex number is its complex conjugate.1 Complex numbers A complex number. the system settles down to its steady state condition. y = Im(z). It follows at once that i2 = −1. We start by summarising the relevant mathematical aspects of complex numbers in Section 3.6. i is called an imaginary number because the square of no real number equals minus one.6. as usual. and the constants B and φ are determined by the initial conditions. while i is defined as the square root of −1: i = −1. by x = Re(z). After an initial transient response.2 we apply complex numbers to the case of forced oscillations with damping. can be written z = x + iy where x and y are real numbers (i. 𝑑2 𝑥2 𝑑𝑥2 2 +𝛾 + 𝜔02 𝑥2 = 0 𝑑𝑡 𝑑𝑡 Hence 𝑑2 (𝑥1 + 𝑥2 ) 𝑑(𝑥1 + 𝑥2 ) 𝐹0 + 𝛾 + 𝜔 2 0 (𝑥1 + 𝑥2 ) = 𝑐𝑜𝑠𝑡 𝑑𝑡 2 𝑑𝑡 𝑚 and so (x1 + x2) is also a solution of the equation in the form.6. which require special provision in engineering design 3. This may produce dangerously high voltages and currents. which is denoted by an asterisk. In Section 3.1. ordinary numbers as we have used so far). respectively. The amplitude A(ω) and the phase angle δ are both functions of driving frequency ω. 𝛾2 1 2 𝑥 = 𝑥1 + 𝑥2 = 𝐴 𝜔 cos 𝜔𝑡 − 𝛿 + 𝐵𝑒𝑥𝑝 −𝛾𝑡 2 cos[(𝜔0 − ) 𝑡 + ] 2 4 for the case of light damping. 3. which is often denoted by z.e.

this corresponds to rotating the line OP through an angle  in the anticlockwise direction to the new position OP’.e. measured in the anticlockwise sense. 3.2 we solved this equation by assuming a solution of the form 𝑥 = 𝐴 𝜔 cos 𝜔𝑡 − 𝛿 .The frequently occurring quantity zz∗. the product of a complex number with its complex conjugate zz∗ = x2 + y2 The below figure is referred to as the Argand diagram of z. as shown in figure.2 Use of the complex representation for forced oscillations with damping The equation for forced oscillations with damping is 𝑑2 𝑥 𝑑𝑥 𝐹0 2 +𝛾 + 𝜔02 𝑥 = 𝑐𝑜𝑠𝑡 𝑑𝑡 𝑑𝑡 𝑚 In Section 3. Similarly multiplying a complex number by eiπ is equivalent to multiplying the number by −1. However. Thus multiplying a complex number by eiπ/2 is equivalent to multiplying the number by i.6. If we multiply reiθ by ei we obtain z’= z ei = rei(θ+) In the Argand diagram. eiπ/2 = cos(π/2) + i sin(π/2) = i.2. and is given by 𝑥 𝑦 𝑐𝑜𝑠𝜃 = . i. If  = π/2 the line is rotated through π/2. The angle θ is the angle that the line OP makes with the positive x direction. y = r sin θ with r being the distance OP: 𝑟 = 𝑥2 + 𝑦2 The real breakthrough comes through employing the important relation due to Euler eiθ = cos θ + i sin θ It follows from this relation we can write z = x + iy = r(cos θ + i sin θ) = reiθ . 𝑠𝑖𝑛𝜃 = 𝑥2 + 𝑦2 𝑥2 + 𝑦2 these relations suggest the introduction of polar coordinates x = r cos θ.

In the complex representation we have the corresponding complex differential equation 𝑑2 𝑧 𝑑𝑧 𝐹0 2 + 𝛾 + 𝜔02 𝑧 = 𝑒 𝑖𝜔𝑡 𝑑𝑡 𝑑𝑡 𝑚 𝑖𝜔𝑡 In particular (F0/m) cos ωt is the real part of (F0/m) 𝑒 . these results have been obtained more readily using the complex representation .and determined the behaviour of A(ω) and δ as functions of ω. We assume a solution of the form z = A(ω)𝑒 𝑖 𝜔𝑡 −𝛿 ) and substitute this in above equation giving 𝐹0 𝑖𝜔𝑡 −𝜔2 𝐴 𝜔 + 𝑖𝛾𝜔𝐴 𝜔 + 𝜔02 𝐴 𝜔 𝑒 𝑖 𝜔𝑡 −𝛿 = 𝑒 . 𝑚 Dividing through by 𝑒 𝑖 𝜔𝑡 −𝛿 we obtain 𝐹0 𝑖𝛿 (𝜔02 − 𝜔2 )𝐴 𝜔 + 𝑖𝛾𝜔𝐴 𝜔 = 𝑒 𝑚 Taking real and imaginary parts of this equation gives 𝐹0 (𝜔02 − 𝜔2 )𝐴 𝜔 = 𝑐𝑜𝑠𝛿 𝑚 and 𝐹0 𝛾𝜔𝐴 𝜔 = 𝑠𝑖𝑛𝛿 𝑚 from which we readily obtain 𝛾𝜔 𝑡𝑎𝑛𝛿 = 2 (𝜔0 − 𝜔 2 ) and 𝐹0 𝑚 𝐴 𝜔 = 1 [ 𝜔02 − 𝜔 2 2 + 𝜔 2 𝛾 2 ]2 These are the same results we obtained in Section 3.2. However.2 using sines and cosines.